The ratio of women to men in India reveals socially determined excess female mortality. The cultural and economic variables used to explain variations in the juvenile sex ratio are interrelated with each other and with other demographic and economic variables, and show large regional differences within India. Our empirical study is built on an extension to Sen's entitlements framework and spatial lag and spatial error econometric procedures. We find that low female labour participation (FLP) is an important determinant of anti-female child bias in regions characterised by both Indo-Aryan and Dravidian kinship systems, but where Indo-Aryan kinship predominates, mainly in Northern India, the effect of FLP is much more significant.